Project Summary DNA copy number alterations (CNAs) are oncogenic drivers for many types of human cancer. For certain decidedly lethal cancers, e.g. high-grade serous ovarian cancer and related forms of breast and endometrial cancers, it is very likely that CNAs rather than point mutations or translocations comprise the bulk of genetic alterations responsible for their highly malignant properties. This may also be true for squamous carcinoma of the lung and for subsets of gastric and esophageal cancers. Despite their deadly impact on hundreds of thousands of cancer patients, relatively little attention is being paid to understanding this class of genetic alterations. More importantly, from the cancer treatment perspective, there is no roadmap for determining whether they induce selective dependencies that could be utilized for developing new therapeutics. The vast majority of CNAs contain multiple driver genes, as our group and others have discovered in the past several years, and this makes it considerably more difficult to study how they impact cancer progression compared to single-gene events. Despite this difficulty, it is important that we continue to make progress. The overall goal of this project is to develop new tools and models to investigate multigenic CNAs so that they can be more readily studied and utilized in developing new therapeutics. In Aim 1 we will combine CRISPR/Cas9 and Cre-loxp genome engineering to model multigenic CNAs accurately and determine how they impact oncogenic phenotypes in normal mammary epithelial cells, similarly to how mutations in single gene alterations such as PIK3CA are currently studied. Once we have validated these new cell models, we will screen for induced dependencies. In Aim 2 we will develop and implement computational methods to extract information about specific CNAs from the vast warehouse of information present in large-scale integrated cancer genome datasets. We have extensive preliminary results that validate this approach, including the prediction of CNA-selective dependencies. Aim 2 will include testing these predictions. Lastly, we believe that true understanding of multigenic CNAs will only come about when we functionally probe the interactions between multiple drivers, which we previously demonstrated was a key feature of the oncogenicity of the multigenic 14q13 amplicon in lung cancer and multigenic 11q13 amplicon in liver cancer. Accordingly, our final key goal is to develop and implement generalizable methods to study genetic interactions between multiple drivers (Aim 3). The clinical effectiveness of targeted treatments for patients with HER2-amplified breast cancers underscores the enormous translational potential of more intensive research into CNAs. By developing the new tools and models for CNAs described in this proposal, we will make a significant impact on understanding multigenic CNAs and laying the groundwork for identifying associated dependencies and therapeutic strategies.